Peak oil: the science behind the problem

Let’s return to the issue of peak oil. In the first post, I explained that fairly soon the world will reach a peak in the amount of oil it can pump out of the ground on a daily basis. Regardless of how much total oil remains in the earth, natural limits exist to how fast we can get the oil out of the ground.

Why? Common sense would seem to dictate that, if the oil is down there, pumping it out at a faster rate is simply a matter of sinking more wells, or building fatter pipelines, or otherwise turning the dial to 11. And common sense would be wrong.

If you’re like me, you might have naively assumed that tapping an oil field was a bit like tapping a beer keg. Once you get the tap in place, you simply pump away until the keg runs dry. All of the glory lies in finding the keg (oil field) in the first place. Afterward, mining the resource is a standard bit of engineering.

In reality, tapping an oil field is a bit like prospecting for beer the morning after a frat party has taken place (warning: strained analogy ahead). You might find a half-empty keg still in its ice bucket, an easy source of fresh beer. Next, you might collect the leftovers in stray cups and cans littering the common areas. This beer would be stale, contaminated, and available in smaller quantities, but still fairly easy to find. Finally you might resort to wringing out the sofa cushions, a time-consuming, thankless process that would yield only dregs.

An oil field works a bit the same way. First you mine the clean, high-pressure pockets, the source of all those gushing oil wells you see in 1950s movies about wildcatters. From there, you go to increasingly desperate lengths to extract lower-quality crude from the intermixed sandstone, shale, clay, and limestone that comprise a typical oil field.

This is the geological constraint that underlies peak oil, and no technology available now or under development will change the incontrovertible fact that an oil field’s daily production rate goes into decline long before its supply of oil is tapped out.

Not only is this process uncontroversial, it’s fairly predictable. Shell geologist M. King Hubbert figured in 1956 that oil production in the continental U.S. would peak in 1970. He was right — daily oil production in the lower 48 has been in decline since December 1970, despite the fact that we have plenty of oil left.

That was 35 years ago. What’s happened since? In a nutshell, the rest of the world’s oil production has peaked too — with the possible exception of the OPEC countries, and Saudi Arabia in particular. The OPEC countries are a bigger mystery. And a subject for a future post.

Again, a major tip of the hat to Kevin Drum for lucidly laying out the peak oil story.

Author Bio


Comments Disabled

  1. Matt Robinson - November 30, 2005

    Nice summary. I think another concrete peice of information is that world consumption is at 84 million barrels a day and rising at a rate of 2 million barrels per day per year. For an excellent support for these numbers and the description of sucking on sofa cushions posted above, read through the attached New York Times Magazine article.
    Also here is a nice place to browse this topic:
    New York Times Magazine article Aug 05:
    This article is only available via the nyt if you have an account, so here is the text:
    Edited for length. To access a version of the article that doesn’t require an account, just follow this link. It’s a great article. Highly recommended.

  2. Adam - November 30, 2005

    Hi Matt,
    Thanks for the link. I hope you don’t mind, but I edited your comment to contain a link to a version of the article that doesn’t require a registration, rather than the full article itself.
    FYI, you can create links to stable, registration-free versions of NYT articles using the NYTimes Link Generator. An indispensible tool for bloggers.